GitError

[minchong1998]

i think your work is charming and i want to learn more about it. when i run the code, it has some error that is:
File "main.py", line 250, in
main(args_)
File "main.py", line 197, in main
checkpoint_saver = Saver(args)
File "/home/mc/Project/stereo-transformer-main/utilities/checkpoint_saver.py", line 24, in init
self.save_experiment_config()
File "/home/mc/Project/stereo-transformer-main/utilities/checkpoint_saver.py", line 41, in save_experiment_config
repo = Repository('.')
File "/opt/anaconda3/envs/GWCNet/lib/python3.6/site-packages/pygit2/repository.py", line 1494, in init
path_backend = init_file_backend(path, flags)
_pygit2.GitError: Repository not found at .

can you give me some advise to deal with it? thank you.

[mli0603]

Hmm, so I am trying to log which git repo the experiment is running by pygit. Apparently the git information is deleted on your end. Can you try the following:

• How did you get the code? Clone? Or download a zip file and then unzip? If you downloaded a zip file, this may be the problem. I recommend you clone the repo if possible.
• If you cloned the repo, can you check if you have a local .git folder inside the stereo-transformer repo folder? If not, this is the cause. I don't know why it happened. Maybe try cloning again will help.
• If you feel you don't need to log the git repo, you can just comment out line 41 and 42 here

[minchong1998]

Thank you, the first advice work! But i get another error like following.Can you help me about this? Thank you very much!

File "main.py", line 250, in
main(args_)
File "main.py", line 221, in main
args.clip_max_norm, amp)
File "/home/mc/Project/stereotrans/utilities/train.py", line 33, in train_one_epoch
_, losses, sampled_disp = forward_pass(model, data, device, criterion, train_stats)
File "/home/mc/Project/stereotrans/utilities/foward_pass.py", line 55, in forward_pass
outputs = model(inputs)
File "/opt/anaconda3/envs/GWCNet/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/mc/Project/stereotrans/module/sttr.py", line 100, in forward
output = self.regression_head(attn_weight, x)
File "/opt/anaconda3/envs/GWCNet/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/mc/Project/stereotrans/module/regression_head.py", line 261, in forward
occ_mask = batched_index_select(occ_mask, 2, x.sampled_cols)
File "/home/mc/Project/stereotrans/utilities/misc.py", line 37, in batched_index_select
return torch.gather(source, dim, index)
RuntimeError: _th_gather not supported on CUDAType for Bool

[mli0603]

Hi @minchong1998, hmm, maybe there is a bug. Can you give me more info? Which branch are you running? Which file gave you this error? There were some updates which may not be consistent across branches.

[minchong1998]

thank you for your reply. I just want to train the model by myself, so I run main.py and it gave me this error. File gave me error is in the error massage.

[mli0603]

• Which branch are you on? Is it main branch?
• Which torch version are you using?

From the log it looks like gather is not supporting bool type. I don't recall an error like this before.

[minchong1998]

Thank you for your reply. It is the torch version.

[mli0603]

Sounds good. Closing this for now.